Automated Platelet-Sizing Parameters on a Normal Population
SUSAN S. GRAHAM, M.S., MT(ASCP)SH, BERNARD TRAUB, M.D., AND IRVING B. MINK, M.S.
The quantitation of platelets in peripheral blood is a well-recognized tool. However, other platelet parameters that recently
have become available on a routine basis with the introduction
of automated cell sizers may become increasingly important in
evaluating the integrity of the thrombocytic function. These parameters include mean platelet volume (MPV) and platelet-crit
(PCT). This study established reliable reference ranges for these
platelet parameters while taking into consideration the effects
of age and sex in a study population of 447 normal persons.
There was an inverse, nonlinear relationship between MPV and
platelet count, with no statistical difference (P > 0.05) seen between males and females. Platelet-crit showed no variation with
respect to age or sex, however, a direct, linear relationship was
suggested between PCT and platelet count. These relationships
provide a better understanding of these platelet parameters and
may contribute to their use as helpful diagnostic aids. (Key words:
Platelets; Platelet size; Mean platelet volume; Platelet-crit) Am
J Clin Pathol 1987; 87: 365-369
THE QUANTITATION of platelets in peripheral blood
is a well-recognized diagnostic tool. However, other platelet parameters that recently have become available on a
routine basis with the introduction of automated cell sizers
may become increasingly important in evaluating the integrity of the thrombocytic function. Mean platelet volume (MPV) has been used to distinguish between thrombocytopenia resulting from peripheral platelet destruction
and that resulting from hypoproliferation.9 This measurement may also be used to evaluate bone marrow
suppression and recovery in chemotherapeutic regimens.5
The MPV is currently available as an adjunct to a routine
complete blood count as performed by an electronic cell
counter.
The Coulter Model S-Plus IV® electronic cell counter
(Coulter Electronics, Hialeah, FL) prepares a histogram
of platelet volumes and from it determines platelet count,
MPV, and platelet distribution width (PDW). This instrument also calculates a platelet-crit (PCT) from the
platelet count and MPV.
Although much work has been performed in the area
of mean platelet volume, some controversy exists as to
the reference range for this parameter. Most studies have
described the reference range for the MPV with respect
to the platelet count, thus inferring that there is no single
normal range for MPV. An acceptable clinical range has
Received March 25, 1986; received revised manuscript and accepted
for publication June 5, 1986.
Address reprint requests to Dr. Traub: Department of Pathology, Millard Fillmore Hospital, 3 Gates Circle, Buffalo, New York 14209.
Department of Pathology, Millard Fillmore Hospitals and the
State University of New York at Buffalo, School of Medicine,
and Department of Natural Sciences, Roswell Park Memorial
Institute Division, Buffalo, New York
not yet been established for the PCT as determined by
the Coulter Model S-Plus IV. The purpose of this study
is to establish tentative reference ranges for the platelet
count, MPV, and PCT in a selected population, while the
effects of age and sex variables are taken into consideration.
Materials and Methods
Instrument and Principles of Operation
The instrument used in this study is the Coulter Model
S-Plus IV, a voltage-pulse automated particle counter and
sizer. Cells are suspended in an electrolyte solution and
are pulled through a small aperture located between two
electrodes. As particles pass through the aperture, an increase in resistance is created and recorded as a voltage
pulse. The size of this pulse is directly proportional to the
volume of the particle. The platelet count is determined
from the number of pulses between 2 and 20 fL in volume.
The data are then fitted to a log-normal distribution if
possible, and the platelet count and mean platelet volume
are derived from the histogram. The PCT is the product
of the MPV (femtoliters) times the platelet count (per
liter) and expressed as a percentage. Quality control was
maintained daily on the instrument, and the means, standard deviation, and coefficient of variation were all within
acceptable limits.
Patient Population and Blood Specimens
The study group used in the determination of normal
range was derived from 477 ambulatory patients undergoing preadmission evaluation for elective surgery before
entering Millard Fillmore Hospital, Buffalo, New York,
betweeen January and June 1984. Millard Fillmore Hospital is an urban facility servicing a large cross-section of
racial and socioeconomic groups. Preadmission testing
included a chemistry screen, complete blood count with
differential, and urinalysis. These results were reviewed
by the authors. The platelet data from all these patients
365
A.J.C.P. • March 1987
GRAHAM, TRAUB, AND MINK
366
12. O
11.0MPV
(fL)
10.0
9. O •
FIG. 1. MPV as a function of time (MPV
± 2 SD).
8. O
7. O
O. O
-r60
-r120
-r180
240
300
TIME
360
-r420
-r480
-V
540
1
24
HOURS
tMIN)
were used with the exception of those results that were
indicative of significant disease or exhibited patterned abnormalities in any of the preadmission screening tests.
The blood samples used in this evaluation were drawn
by venipuncture into evacuated 5-mL potassium-EDTA
anticoagulated sterile tubes (Vacutainer®, Becton-Dickinson, Rutherford, NJ) and were analyzed for the ordered
presurgical complete blood counts. The platelet study data
(except for platelet count and MPV) were neither reported
nor made part of the patients' medical record. In addition,
because the PCT and PDW are not FDA approved, written release was obtained from Coulter Electronics, stating
that these values were to be used solely for research purposes.
cade and by sex, with the exception of pediatric patients,
who were grouped according tofive-yearintervals. If no
significant differences were found, categories were combined to achieve more meaningful results. A reference
range was determined with the use of all of the data collected. A mean and standard deviation were generated,
and the "normal" range was defined as the mean ± 2
standard deviations. Tests for statistical significance were
performed with the use of t-test analysis. Any patterned
relationships were analyzed.
Methods
The mean platelet volume (MPV) increased 12% in the
first 15 minutes after exposure to EDTA (Fig. 1). An additional 6% increase was observed during the remainder
of the first hour. Between one hour and six hours, the
MPV remained relatively constant with only a 3% increase
noted. This represents a change in platelet volume of 0.3
fL. In the specimens refrigerated for 24 hours, there was
an 8% increase (0.7 fL) in MPV as compared with the
values obtained after 60 minutes.
An 8% increase in the PCT was found during the first
15 minutes, and and additional 5% increase was observed
in the next 45 minutes. Between 1 hour and 24 hours,
the PCT remained relatively constant with only a 4% increase (0.009%).
The parameter most significantly affected by time and
exposure to anticoagulant was the MPV. Because of these
results, all subsequent samples to be used in this evaluation
were tested between one and six hours after venipuncture.
Initial Study. Previous reports have indicated that anticoagulated whole blood should be allowed to stand at
room temperature for a minimum of two hours so that
stability of platelet parameters can be obtained.7 This is
because of the change in shape, from discoid to spheric,
that platelets undergo when exposed to EDTA. Because
an exact two-hour wait is not always feasible in a busy
laboratory, changes over time were studied. Whole blood
samples from 16 normal people were analyzed at 0, 15,
30, 45, 60, 120, 240, and 360 minutes to determine the
percentage of change over time in each parameter to be
studied. Additionally, the samples were then refrigerated
at 4-5 °C and were analyzed at 24 hours after returning
to room temperature to evaluate the effect of refrigeration.
This was done to ascertain whether whole blood samples
refrigerated overnight are acceptable for analysis the following day.
Subsequent Studies and Data Reduction. After the optimal time for testing was determined, the remainder of
the patients were studied over a six-month period. Initially,
the patients' results were grouped by age according to de-
Results
Effect ofAnticoagulant and Time
Platelet Count
Tables 1 and 2 represent the results for all the platelet
parameters studied. The mean platelet count for the entire
AUTOMATED PLATELET-SIZING PARAMETERS
Vol. 87 • No. 3
Table 1. Platelet Parameters in Males (mean ± SD)*
Age
(years)
1-5
6-10
11-15
16-20
21-30
31-40
41-50
51-60
61-70
71-86
MPV
(fL)
PCT
n
Platelet Count
(X103M-)
(X10'/L)
24
24
16
16
24
12
17
22
29
23
357 ±70
351+85
282 ± 63
266 ±63
238 ±49
244 ± 56
271 ±66
258±61
256 ± 53
237 ±49
8.6 ±0.7
8.6 ±0.8
9.8 ±1.0
10.2 ±1.1
9.6 ±0.6
9.8 ±1.2
9.4 ±1.0
9.8 ±1.2
9.4 ±1.1
9.6 ±1.0
367
Table 2. Platelet Parameters in Females (mean ± SD)
(%)
Age
(years)
0.304 ±0.059
0.300 ±0.058
0.274 ±0.053
0.266 ±0.049
0.277 ± 0.045
0.237 ±0.044
0.250 ±0.045
0.248 ±0.045
0.238 ±0.047
0.226 ±0.048
1-5
6-10
11-15
16-20
21-30
31-40
41-50
51-60
61-70
71-83
MPV
(fL)
PCT
n
Platelet Count
(X103ML)
(X10'/L)
25
18
31
22
43
30
26
21
30
24
381 ±76
336 ± 76
298 ± 72
270 ±58
270 ±58
282 ± 56
279 ±65
285 ± 54
274 ±61
279 ± 65
8.9 ±0.8
9.7 ±1.1
9.8 ±1.2
9.7 ±0.7
9.8 ±1.0
9.8 ±1.2
9.8 ±0.9
9.7 ±0.7
9.6 ±0.9
9.5 ±1.0
0.337 ±0.069
0.326 ±0.080
0.288 ±0.058
0.262 ±0.058
0.261 ±0.046
0.271 ±0.046
0.274 ±0.072
0.276 ±0.045
0.262 ±0.052
0.261 ±0.054
(%)
* SI conversion units for platelet count X10 J /ML is platelet count XI0 9 /L in Figures 2, 3,
and 4.
group of 477 people, ranging in age from 1 to 86 years,
was 286,200/ML (286 X 109/L). Based on the information
in Tables 1 and 2, the groups were combined to form six
divisions according to age and sex (Table 3), in order to
allow for clinically useful ranges.
Mean Platelet Volume
The results are summarized in Tables 1 and 2. There
was no evidence of a patterned relationship when different
ages and sexes were compared. Differences by age appear
to relate to platelet count by age. The combined sample
mean was 9.6 ±2.1 fL. When the MPV was compared
as a function of platelet count (Fig. 2), there was found
to be an inverse, nonlinear relationship between the two
parameters.
Platelet-Crit
The information gathered for PCT is presented in Tables 1 and 2. The ranges for males and females in each
of the age groups remained relatively stable. There was a
direct, approximately linear, relationship noted between
PCT and platelet count (Fig. 3).
frigeration, samples for platelet volume analysis should
not be refrigerated overnight, as is common practice for
routine complete blood counts.
Platelet count seemed to vary somewhat with age. This
study suggests the use of two reference ranges: pediatric,
including both males and females, ages 1-15 years, and
adult, including members of both sexes older than 15 years
of age. The pediatric range would be 335,600 ± 161,800/
nh (336 ± 162 X 109/L). The adult range would be
266,100 ± 1 14,700/ML (266 ± 115 X 109/L). The results
from males and females, ages 11-15 years, were assigned
to the pediatric group, however, they might be better
treated as a separate group or possibly reassigned by individual yearly categories to the appropriate age grouping.
Adult females, older than 20 years, exhibited consistently
higher platelet counts when compared with males of the
same age group, however, this difference was not thought
to be clinically significant.
There was an inverse nonlinear relationship between
MPV and platelet count. This agrees with previous findings, which suggest the use of a nomogram to interpret
mean platelet volume as a function of platelet count.3
There was no statistically significant difference seen between males and females older than 15 years of age. In
females 1-15 years of age, the MPV tended to be higher
than in males of the same age group. Statistically, this
Discussion
With the exception of platelet count, there were significant changes in the platelet parameters in the first 60
minutes after venipuncture. After thefirsthour, the MPV
and PCT remained relatively stable for at leastfivehours.
This agreed with a study by Bessman,1 which reported a
20% increase in volume shape change in the first hour,
after which the MPV was stable for at least 12 hours, and
disagreed with reports by Marugan and associates8 and
Small and Bettigole,12 who found the change in the platelet
volume not complete until two hours after venipuncture.
Because the MPV increased 8% during the 24-hour re-
Table 3. Platelet Count/Microliter (X106/L) in Normal
Persons (mean ± 2 SD)
Age (years)
Male
Female
Male and female
1-15
16-86
n = 64
335,900 ±160,000
n = 74
335,500 ± 163,300
n = 138
335,600+161,800
n = 143
252,100 ± 112.600
n=196
276,200 ± 116,200
n = 339
266,100 ± 114,700
MOLE
FEMALE
MPV
(n)
FIG. 2. MPV as a function of platelet count
(MPV ± 2 SD).
101-150
151-200
201-250
251-300
Platelet
301-350
Count
351-400
401-450
451-500
501-550
(X103/ul)
MRLE
FEMRLE
•
T
T
PCT
T
l
•
IX)
-
T
T
t
1
J.
5
t
t
1
T
1
t
1
I
1
1'
,
1
T
1
1
1
l
I
I
T
1
i
1
3
FIG. 3. PCT as a function of platelet count
(PCT ± 2 SD).
i
L
1
i
- —
"
i
r "
•
101-150
151-200
201-250
"i1
•'•'
251-300
Platelet
'
301-350
Count
1
1
351-400
401-450
1
451-500
501-550
(X1Q3/ul)
PCT
(X)
FIG. 4. PCT as a function of platelet count
with upper and lower reference limits.
101-150
151-200
201-250
251-300
Platelet
301-350
Count
351-400
401-450
3
(X10 /ul)
451-500
Vol. 87 • No. 3
AUTOMATED PLATELET-SIZING PARAMETERS
difference was significant, however, larger samples should
be tested to verify this finding.
PCT showed no significant variation in males or females
over the range of ages studied. Plotting mean PCT (%)
versus platelet counts between 100 and 500 (X103)/^L
(X109/L) (Fig. 4), a direct, linear relationship was suggested between PCT and platelet count such that Y
= (6.96 X 10"4)X + 0.06, where Y is PCT in percent and
X is platelet count X10 3 /ML (X109/L). The correlation
coefficient for this line was 0.998. The mean standard
deviation was used in developing the reference range,
which would be represented by the PCT derived from the
equation using the appropriate platelet count ± 0.054. In
certain laboratory situations, a mathematical interpretation might be advantageous. In the normal population
used in this study, there were insufficient numbers of
samples whose platelet counts were in the higher ranges.
Future studies should incorporate a greater number of
samples in these ranges to refine the relationship.
The stability of these parameters in the normal population helps to distinguish platelet abnormalities when they
exist. The potential diagnostic value of these parameters
is currently being studied by many authors. It has been
demonstrated that patients with thrombocytopenia resulting from peripheral destruction have an increased
MPV.2 This is in contrast to those patients whose thrombocytopenia is hypoproliferative in nature where the MPV
is decreased.5 This is an important differentiation because
patients with hypoproliferative platelet disorders have a
greater tendency to bleed. Those patients whose thrombocytopenia results from increased destruction have a low
platelet count, but a greater percentage of those platelets
369
are younger, larger, and have an increased functional capacity.10 These parameters are also being studied as diagnostic indicators in myeloproliferative disease,11,12 hypersplenism,6 and megaloblastic anemia.4 The PCT may
add to these parameters as a useful diagnostic aid.
Acknowledgments. Gary Stone and Frederick Anner provided the
graphics and computer analyses.
References
1. Bessman JD: New parameters on automated hematology instruments. Laboratory Medicine 1983; 14:488-491
2. Bessman JD: The relation of megakaryocyte ploidy to platelet volume. Am J Hematol 1984; 16:161-170
3. Bessman JD, Levin J: The inverse relation of platelet size and count
in normal subjects and as an artifact of other particles. Am J Clin
Pathol 1981;76:289-293
4. Bessman JD, Williams LI, Gilmer PR: Platelet size in health and
hematologic disease. Am J Clin Pathol 1982; 78:150-153
5. Fishledger AJ, Hoffman GC: Automated hematology: counts and
indices. Laboratory Management 1984; 22:21-34
6. Karpatkin S, Freedman L: Hypersplenic thrombocytopenia differentiated from increased peripheral destruction by platelet volume.
Ann Intern Med 1978; 89:200-203
7. Levin J, Bessman JD: The inverse relation between platelet volume
and platelet number. J Lab Clin Med 1983; 101:295-307
8. Marugan I, Sabater V, Sanjuan L, Garcia-Conde J: Evaluation of
platelet parameters obtained with the Coulter S Plus in a normal
population. Sangre 1983; 28:512-522
9. Nelson B, Kehl D: Electronically determined platelet indices in
thrombocytopenic patients. Cancer 1981; 48:954-956
10. Paulus J: Platelet size in man. Blood 1975; 46(3):321-336
11. Roper PR, Johnston D, Austin J, Agarwal SS, Drewinko B: Profiles
of platelet volume distributions in normal individuals and in
patients with acute leukemia. Am J Clin Pathol 1977; 68:449457
12. Small BM, Bettigole RE: Diagnosis of myeloproliferative disease by
analysis of platelet volume distribution. Am J Clin Pathol 1981;
76:685-691